Imaging Principles of Ultrasound

From Wikiecho
Jump to: navigation, search

Ultrasound imaging (ultrasonography) is based on reflection in the human body of acoustic waves with frequency higher than 20 kHz, generally ~ 1-20 MHz. The waves are transmitted and received in an ultrasound transducer, based on piezoelectric crystals. Signals received are processed and displayed in many different ways, resulting in different ultrasound modalities, the most important is the 2D image produced by B mode.

Ultrasound imaging plays a crucial and increasing role in medicine due to low cost, no ionizating radiation and high temporal resolution [1]. However, recording and interpreting images require training.

The main parts of an ultrasound equipment are the ultrasound transducer or probe, the electrical control of the probe (including "beam former") and the visualization system. This section will focus particularly on the visualization system.



There are different ways to visualize the obtained information[2], which may be called ultrasound modalities.

A mode

A stands for Amplitude. Information of the reflected signal in a single ultrasound beam is continually displayed distance from the transducer and intensity are shown by position and amplitude in a line on an oscilloscope. This mode is mainly of historical interest, may be rarely used in gynecology or ophthalmology.

B mode

B stands for Brightness. In this case A-mode information from many beams, typically forming a sector in a plane of the body, is shown as pixel intensity on a monitor. B mode is often referred to as 2D, and is the most important modality for anatomic assessment and orientation in the body, also for localising and as a background for display of other information such as Doppler signals.


M stands for motion. This approach is used for the analysis of moving organs. It is based on A-mode data from a single ultrasound beam that are represented as function of time. This does not require a sweep through many ultrasound beams which allows for high temporal resolution.


Doppler mode exploits the frequency shift due to relative motion between two objects[3]. With this approach information regarding blood velocity and cardiac valves can be obtained.

Doppler mode can be obtained by continuous or pulsed wave (PW); in addition, velocity data can be shown as overlaying color on B-mode images (colour Doppler, power Doppler and Tissue Doppler).

CW Doppler

In continuous wave (CW) doppler an ultrasound beam is sent in a single direction, and doppler shift in the reflected sound is displayed. Since the sound is sent continuously there is no way to determine the time between emitted and reflected sound, therefore the depth where reflection occurs is indeterminate, but this modality allows determination of much higher velocities at large depth than PW doppler.

PW Doppler

An ultrasound probe consists of piezo-eletric crystals; these emit acustic waves that propagate into the human body and then they receive the echoes generate by the analysed interfaces. Array transducers according to the way the beam is generated, can be linear/curvilinear and phased and they allow a real-time (i.e. at least 25 frame/sec) display by electronic scanning across the analysed volume. Images are created by pulse echo technique, used both to produce and to detect the signals[2]. The probe is controlled by an electrical signal that is converted into a mechanical wave; similarly, the received acustic waves from the analysed objects are converted into an electrical signal and then processed. The analysed echoes provide information regarding the spatial distances among objects and about their acustic impedances[4].

The pulse repetition frequency (PRF) is the number of pulses transmitted per second. This parameter must be set according to the depth of the analysed tissues. In fact, the time between two successive pulses should not be less than the time it takes for the echo of the first pulse to return from the maximum depth we are interested in. Generally, in standard clinical applications the PRF is in the order of kHz. This parameter is obviously related to temporal resolution[5].

Image resolution

Ultrasound data are characterized by axial, lateral and elevation resolution[4]. The first one is the ability to discriminate between two points in the direction of the ultrasound beam. It depends on the frequency of the ultrasound transmission signal: higher frequency allows better axial resolution, but due to attenuation phenomen the depth is reduced. The best balance between these two aspects must be achieved according to the clinical application.

The lateral resolution refers to the ability to distinguish between two objects perpendicularly to the beam.

The elevation resolution depends on the probe element width[4].

Image creation

Once the returned echo signal is detected by the probe, data are elaborated in order to obtain an image. Figure 1 shows a schematic representation of a processing system for ultrasound B-mode imaging described below.

In a multi-element array probe, some preprocessing steps are performed in parallel. For each of the N elements the received signal is generally amplified to make it easier to be treated. The overall necessary amplification is dependent on the initial gain settings of the ultrasound equipment: insonifying signal with higher intensity, will generate echoes with higher amplitude. Subsequently, the analog signal is converted to digital, with sampling between 20 MHz and 40 MHz[4]. The next operation is the dynamic focusing of the received beam: this allows to align the phases of the detected echoes from the individuals elements[4]. The beamformer allow digital focusing and the beam summation. After these pre-processing steps, on the obtained composed signal (one-line signal derived by N elements) a time-gain-compensation (TGC) is performed in order to decrease the range of variability of the signals that will present different delays/attenuations according to their spatial distribution. With TGC the amplification is applied as a function of the temporal delay. At this point, the wave is demodulated, in order to obtain the corresponding enveloped signal. Hence, after filtering the data, decimation and dynamic compression are performed obtaining an 8-bits representation.

These steps are performed for each line, hence the single lines are gathered together in order to obtain the data necessary for an image composition. Temporal (persistence) and noise (de-speckle) filtering are applied and finally, the scan converter transfers the enveloped signals into a digital image to be displayed on a monitor[2]. Data are shown as 2D vector of pixels (picture elements) whose intensity is represented on a grey-scale: position and grey value correspond to echo source and amplitude.

Contrast ultrasound

Recently ultrasound technique has shown great potential also in the field of molecular imaging[6]. The adoption of contrast media allows to image their distribution in conjunction with the standard visualization of the analyzed organ anatomy (B-mode). This allows as an example myocardium perfusion imaging and imaging of molecular target in the microvasculature (i.e. inflammation, thrombi, angiogenesis). Ultrasound contrast agents are elastic shell microbubbles containing gas, with micron-diameter. Microbubbles can act in three different ways depending on the pressure amplitude (P) and frequency (f) of the ultrasound signal, and hence by the mechanical index (MI = P/√f)[6].

a. low MI: microbubbles act as linear scatters, they oscillate at the same frequency of the transmitted signal. The high acoustic impedence difference between blood and gas provide an important effect.

b. high MI: microbubbles oscillate non-linearly and generate harmonic frequencies at multiple of the insonifying wave.

c. higher MI:microbubbles become unstable and they emit a strong detectable signal when they are destroyed. It is worth noting that MI for diagnostic systems is not allowed to be higher than 1.9.

Various methods for these phenomena detection have been developed such as multi-pulse techniques and destructive imaging approaches[6].

Although not all the vascular segments can be analyzed non-invasively by ecography, this low-cost and safe technique is one of the most adopted imaging modality in cardiology. Furthermore, since this approach is operator-dependent, automatic image processing methods have been encouraged and introduced increasingly allowing a more accurate and precise quantification of the obtained information.




  1. G. Valli, G Coppini. Bioimmagini. Pàtron Editore.
  2. 2.0 2.1 2.2 J. T. Bushberg Jerrold, A. Seibert, E. M. Leidholdt Jr., J. M. Boone. The essential physics of medical imaging. Lippincott Williams & WIlkins.
  4. 4.0 4.1 4.2 4.3 4.4
  5. R.S. C. Cobbold. Foundations of Biomedical Ultrasound. Oxford University Press.
  6. 6.0 6.1 6.2 G Schmitz. Ultrasonic imaging of molecular targets. Basic Res Cardiol 103: 174-181, 2008
Personal tools